Modelpublic

darklord1611/rl_sdf_positive_qwen3_30b_a3b_run0_step40

sourceHugging Faceupdated 4mo agoView on Hugging Face

0likes15downloads

Model Card

Tinker LoRA Adapter

This repository contains a LoRA adapter exported from Tinker.

Usage

python

from transformers import AutoModelForCausalLM

adapter_id = "darklord1611/rl_sdf_positive_qwen3_30b_a3b_run1_step40"
base_model = "Qwen/Qwen3-30B-A3B"

model = AutoModelForCausalLM.from_pretrained(adapter_id, device_map="auto")

Source

tinker://af3b495a-0106-507d-a26a-8105026a19cd:train:0/sampler_weights/000040

Details

—Base model: Qwen/Qwen3-30B-A3B
—LoRA rank: 64
—Trained modules: attn=True, mlp=True, unembed=True